EPA/540/R-94/513
ACID EXTRACTION TREATMENT SYSTEM
FOR TREATMENT OF METAL CONTAMINATED SOILS
Prepared by
Stephen W. Paff, Brian Bosilovich, and Nicholas J. Kardos
Center for Hazardous Materials Research
320 William Pitt Way
Pittsburgh, PA 15238
Contract No:. CR-815792-01-0
Project
Kim Lisa Kreiton
Super-fund Technology Demonstration Division
Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
Rl S K REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
NOTICE
This material has been funded wholly or in part by the United States Environmental
Protection Agency under Contract CR-815792-01-0 to the Center for Hazardous
Materials Research. It has been subject to the Agency's review and it has been
approved for publication as an EPA document. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
-------
FOREWORD
The Superfund innovative Technology Evaluation (SITE) Program was authorized in the
1986 Superfund Amendments. The program is a joint effort between EPA's Office
of Research and Development and Office of Solid Waste and Emergency Response.
The purpose of the program is to assist the development of hazardous waste
treatment technologies necessary to implement new cleanup standards which require
greater reliance on permanent remedies. This is accomplished through technology
demonstration designed to provide engineering and cost data on selected
technologies.
The Risk Reduction Engineering Laboratory (RREL) is responsible for planning,
implementing and managing research, development, and demonstration programs to
provide an authoritative, defensible engineering basis for support of the policies,
programs, and regulations of the EPA. This publication is one of the products of that
research and provides a vital communication link between the research and the user
community.
This project consisted of demonstration of the Center for Hazardous Materials
Research's (CHMR) Acid Extraction Treatment System (AETS), which is a process for
the treatment of soils contaminated with heavy metals. The project included the
development of the process from laboratory-scale proof-of-concept testing through
batch pilot-scale tests. AETS was tested using both surrogate soils and soils removed
from current Superfund sites. These soils were contaminated with a variety of
metals, including lead, cadmium, chromium, nickel, arsenic, copper, and zinc. The
goals of the study were to develop and evaluate AETS.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
-------
ABSTRACT
Through a Cooperative Agreement with the U.S. Environmental Protection Agency's
Risk Reduction Engineering Laboratory, the Center for Hazardous Materials Research
(CHMR) developed the Acid Extraction Treatment System (AETS). The project was
conducted with support from Interbeton bv and The Netherlands Organization for
Applied Scientific Research (TNO), located in the Netherlands. AETS is intended to
reduce the concentrations and/or leachability of heavy metals in contaminated soils
to render the soils suitable to be returned to the site from which they originated.
Additional applications may include treatment of contaminated sediments, sludge and
other heavy metal-containing solids.
The objective of the project was to determine the effectiveness and commercial
viability of the AETS process in reducing the concentrations and leachability of heavy
metals in soils to acceptable levels. This report represents an account of the activities
conducted during the project, the experiments performed, and the results.
Five soils were tested, including EPA Synthetic Soil Matrix (SSM), and soils from four
Superfund sites (NL Industries in Pedricktown, NJ; King of Prussia site in Winslow
Township, NJ; smelter site in Butte, Montana; and Palmerton Zinc site in Palmerton,
PA). These soils contained elevated concentrations of arsenic, cadmium, chromium,
copper, lead, nickel, and zinc.
The results of the study are summarized below:
» AETS is capable of treating a wide range of soils, containing a wide
range of heavy metals to reduce the TCLP below the RCRA limit and to
reduce the total metals concentrations below the California-mandated
total metals limitations.
In most cases, AETS is capable of treating the entire soil, with no
separate stabilization and disposal for fines or clay particles, to the
required TCLP and total limits. The only exception to this among the
soils tested was the SSM, which may require separate stabilization and
disposal of 20% of the soil because of lead. SSM was successfully
treated for other metals, including arsenic, cadmium, chromium, copper,
nickel and zinc. A modular system design will allow for the required
flexibility to treat a range of soils.
* Costs for treatment, under expected process conditions, range between
$100 and 180 per cubic yard of soil, depending on the site size, soil
types and contaminant concentrations. Operating costs ranged between
$50 and 80 per cubic yard.
-------
TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS vi
1.0 INTRODUCTION 1
1.1 OBJECTIVE OF THE PROJECT.. 1
1.2 PROCESS DESCRIPTION 1
1.3 PROJECT TEAM 2
1.4 SUMMARY OF LABORATORY SCALE EXPERIMENTS (FIRST YEAR) 3
1.4.1 Laboratory Scale Experimental Procedure 4
1.4.2 Laboratory Scale Results 4
2.0 SUMMARY OF SECOND YEAR ACTIVITIES ,. 6
2.1 SUMMARY OF EXPERIMENTAL PROCEDURES 7
2.1.1 First Extraction Procedure 8
2.1.2 Procedure Changes for Second Set of Extractions 9
2.1.3 Procedure Changes for Final Extractions 9
2.1.4 Laboratory Analytical Procedures 11
2.2 SUMMARY OF PILOT-SCALE EXPERIMENTS 12
3.0 RESULTS AND DISCUSSION 14
3.1 TREATMENT GOALS 14
3.2 BUTTE, MONTANA SOIL 15
3.3 KING OF PRUSSIA, NJ SOIL 18
3.4 SYNTHETIC SOIL MATRIX 23
3.5 PEDRICKTOWN, NJ SOIL 28
3.6 PALMERTON, PA SOIL 31
3.7 ACID REGENERATION 33
3.8 SOIL POST-TREATMENT 34
3.9 FINES TREATMENT 34
4.0 CONCLUSIONS 36
4.1 SOIL AND METAL TREATABILITY 36
4.2 AETS PROCESS DESIGN 37
5.0 AETS ECONOMICS 43
5.1 COST CALCULATIONS 43
5.2 COST SUMMARY 44
APPENDIX A . EQUIPMENT LISTS 46
APPENDIX B . DATA QUALITY CONTROL 50
IV
-------
LIST OF TABLES
TABLE
TITLE
Page
1 SARM Laboratory Scale Extraction 4
2 Total and TCLP Metal Treatment Requirements 14
3 Butte, Montana, Screen Analysis 15
4 AE-102: Butte Soil, 1st Experiment 17
5 AE-104: Butte Soil, 3rd Experiment 17
6 AE-112: Butte Soil, 4th Experiment 17
7 AE-119: Butte Soil, 6th Experiment 18
8 King of Prussia, NJ, Screen Analysis 19
9 AE-115: King of Prussia Soil, 1 st Experiment 20
10 AE-116: King of Prussia Soil, 2nd Experiment 20
11 AE-118: King of Prussia Soil, 3rd Experiment 21
12 AE-122: King of Prussia Soil, 4th Experiment 21
13 Overall King of Prussia Results with Comparison to Heidimij Results .... 22
14 Synthetic Soil Matrix Screen Analysis 23
15 AE-105: Synthetic Soil Matrix, 1st Experiment 25
16 AE-106: Synthetic Soil Matrix, 2nd Experiment 25
17 AE-120: Synthetic Soil Matrix, 3rd Experiment 26
18 AE-121: Synthetic Soil Matrix, 4th Experiment 26
19 SSM Soil Composite Results Using Data from AE-120 and AE-121 .... 27
20 Pedricktown, NJ, Screen Analysis 28
21 AE-107: Pedricktown Soil, 1st Experiment 30
22 AE-114: Pedricktown Soil, 2nd Experiment 30
23 AE-108: Palmerton Soil Experiment 32
24 Regeneration System Metals Removals for AE-105 33
25 Regeneration System Metals Removals for AE-107 33
26 Qualitative Results of Extractions 36
27 AETS Cost Summaries Under Various Conditions 43
LIST OF FIGURES
FIGURE TITLE
1 AETS Block Flow Diagram 2
2 Extraction Flow Diagram 8
3 Revised AETS Flow Diagram 10
4 Dewatering Flow Diagram 11
5 Rinsing Flow Diagram 12
6 Butte Soil Particle Size Distribution 16
7 King of Prussia Particle Size Distribution 19
8 Synthetic Soil Matrix Particle Size Distribution 24
9 Pedricktown, NJ, Soil Particle Size Distribution 29
10 Palmerton Soil Particle Size Distribution 31
11 AETS Pre-Treatment 38
12 Extraction System 39
13 Coarse Solids Dewatering and Rinsing System 40
14 Fines Dewatering and Rinsing 41
-------
ACKNOWLEDGEMENTS
The Center for Hazardous Materials Research (CHMR) would like to acknowledge the
cooperation received throughout the project from the U.S. Environmental Protection
Agency, Office of Research and Development. In particular, we would like to thank
Ms. Kim Lisa Kreiton, from the EPA's Superfund Technology Demonstration Division
who provided helpful guidance on the project. CHMR would like to acknowledge the
following individuals from the research team who were responsible for important
contributions to this project.
Center For Hazardous Materials Research
Mr. Stephen W. Paff Project Manager
Mr. Brian Bosilovich Project Engineer
Mr. Steve Deppen Research Associate
Mr. Nick Kardos Laboratory Technician
The Netherlands Organization For Applied Scientific Research (TNO)
Mr. Karel De Waal Project Director
Mr. Jan Willem Assink Project Engineer
Interbeton bv
Mr. Henri de Laat Project Manager
Mr. Ben Spruijtenburg Project Engineer
Work on this project was funded through Cooperative Agreement Number CR-
815792-01-0 established between the U.S. Environmental Protection Agency, Office
of Research & Development, and the Center for Hazardous Material Research. The
organizations which contributed to funding under this cooperative agreement include
the EPA, CHMR, Interbeton and TNO. The members of the project research team
appreciate the opportunity to participate in this important project to research and
develop a new and innovative technology for the remediation and treatment of
contaminated soils. Inquiries concerning this report, the project or the AETS
technology may be addressed to:
Mr. Stephen W. Paff, CHMM
Manager, Technology Development
Center for Hazardous Materials Research
320 William Pitt Way
Pittsburgh, PA 15238
(412) 826-5320
VI
-------
1.0 INTRODUCTION
Through a Cooperative Agreement with the U.S. Environmental Protection Agency's
Risk Reduction Engineering Laboratory, the Center for Hazardous Materials Research
(CHMR), developed the Acid Extraction Treatment System (AETS), which can be
used to remove heavy metals from soils, solids, and sludge. The project was
conducted with support from Interbeton bv and The Netherlands Organization for
Applied Scientific Research (TNO), located in the Netherlands. CHMR, a not-for-profit
subsidiary of the University of Pittsburgh Trust, is located in Pittsburgh, Pennsylvania.
This project represents an extension of existing "soil washing" techniques in the
Netherlands, which have been directed primarily at the remediation of hazardous
organic contamination. AETS is intended to treat soils contaminated with heavy
metals. Additional applications may include treatment of contaminated sediments,
sludge and other heavy metal-containing solids.
1.1 OBJECTIVE OF THE PROJECT
The objective of the project was to determine the effectiveness and commercial
viability of the AETS process in reducing the concentrations and/or leachability of
heavy metals in soils to acceptable levels as defined by Federal and state (particularly
California) regulations. This was to be accomplished by a combination of laboratory-
scale investigations during the first year (to prove the concept of acid extraction and
establish process parameters) followed by design, construction and testing of a pilot-
scale extraction unit during the second year. The process viability and economics
were to be determined based on the results of the pilot-scale experiments.
1.2 PROCESS DESCRIPTION
A simplified block flow diagram of the AETS process is given in Figure 1. The first
step in the AETS process is screening to remove coarse solids. These solids, typically
greater than 4 mm in size, are relatively clean, requiring at most a simple rinse with
water or detergent to remove smaller attached particles.
After coarse particle removal, the remaining soil is scrubbed in an attrition scrubber
to break up agglomerates and cleanse surfaces. Then it is contacted with acid (HCI)
in the extraction unit. The residence time in the unit will vary depending on the soil
type, contaminants and contaminant concentrations, but generally ranges between 10
and 40 minutes. The soil/extractant mixture is continuously pumped out of the mixing
tank, and the soil and extractant are separated using hydrocyclones. The solids are
piped to the rinse system, while the extractant is treated using a proprietary
technology which removes the metals and regenerates the acid. The soils are rinsed
with water to remove entrained acid and metals. CHMR anticipates a final step, not
currently performed, in which the soils will be mixed with lime and fertilizer to
neutralize any residual acid and return the soil to natural conditions.
-------
CONTAMINATED
SOIL
CLASSIFICATION
(SCREENING)
COARSE SOIL
PARTICLES
IAKE-UP
ACID ^-
RINSE
WATER >~
EXTRACTION
UNIT
t
RINSE /
\
DEWATER
r
NEUTRALIZATION
& STABILIZATION
REGENERATED ACID
EXTRACTANT
^
^
RlNSATE ^^
ENTTRA1NED
^ SOILS
i
ACID
REGENERATION,
1 i
^ T
HEAVY METALS
>- TREATED SOIL
Figure ^ AETS Block Flow Diagram
1 . 3 PROJECT TEAM
The AETS research and development project is principally supported by an
interdisciplinary project team of individuals from the following four organizations
recognized internationally as leaders in the research, development and
commercialization of environmental technologies.
The Center for Hazardous Materials Research (CHMR), a non-profit subsidiary of the
University of Pittsburgh Trust, was formed in 1985 to assist industry and government
develop practical solutions to the many problems associated with the use and disposal
of hazardous materials and solid wastes. CHMR pursues its mission by conducting
broadly based interdisciplinary applied research, technical assistance, education, and
public policy programs on issues involving hazardous materials and waste in
partnership with academic, industrial, and government organizations. CHMR is active
in fostering international research and technology transfer, and has been providing
research and development capabilities to businesses and industry, including assisting
in the commercial development of new and innovative technologies.
The National Environmental Technology Applications Corporation (NETAC) is a unique
public-private joint venture created in 1988 by the U.S. Environmental Protection
Agency and the University of Pittsburgh Trust. Its purpose is to facilitate the
commercialization of innovative environmental technologies that may positively impact
the nation's most pressing environmental problems. NETAC provides assistance in
the transfer of new environmental technologies from government, university, and
private sector laboratories to the marketplace through a flexible program of technology
-------
evaluations, applied research, bench and pilot-plant testing, and demonstration
projects.
Both CHMR and NETAC operate out of the University of Pittsburgh's Applied Research
Center (formerly Gulf Oil's R&D Center), which provides a combination of both
technical expertise and superior applied research facilities.
The Netherlands Organization For Applied Scientific Research (TNO) is among the
leading applied research institutions in the world. With a staff of 5,000, TNO
researches, develops, and supplies state-of-the-art technology as well as innovative
new technologies. TNO performed the initial work on AETS and played an important
role in working with Dutch industry in the successful commercialization of a soil
washing process for organic contaminants. TNO has licensed numerous technologies
that are now commercially used throughout the world.
Interbeton bv is the international operating company of Hollandsche Beton Groep nv
(HBG), founded in 1902. The HBG subsidiaries comprise nine operating companies:
each with its own particular specialization. Since its establishment in 1958,
Interbeton has gained broad international recognition, experience, and know-how;
particularly in the field of general civil engineering and harbor construction. Recent
Interbeton projects include the Dammam Port Development and King Khaled City
Building project in Saudi Arabia; construction of the Hampton Road Tunnel in Virginia;
completion of the Boston Harbor risers project; and construction of the Boston Harbor
crossing tunnel. Interbeton is active and experienced in combining the know-how
existing within HBG to research and develop new and innovative advanced equipment
for civil construction; including the successful research, development, and
commercialization of soil washing technology in the Netherlands.
1.4 SUMMARY OF LABORATORY SCALE EXPERIMENTS (FIRST YEAR)
The main goal of the first-year laboratory scale experiments was to determine the
overall viability of the AETS process. CHMR also tried to identify potential problem
areas which might occur in subsequent work. The following is a list of activities
which were accomplished and the results of the laboratory scale experiments.
The optimum pH to perform extractions was found to be between 1.5 and
3.0, based on experiments at different pH levels.
Preliminary flow diagrams were developed and evaluated for the AETS
process. These diagrams were used to design and build the pilot-scale
extraction unit.
The amount of acid used in the extraction is dependent on the type of soil
being treated. Some soils, such as the Synthetic Soil Matrix (SSM), have
very good buffering ability and therefore require more acid.
-------
The rinsing stage was found to be a critical step in the process design.
The metals which are transferred from the soil to the extractant (acid)
solution are removed during the rinsing. More efficient rinsing removes
more of the metals.
1.4.1 Laboratory Scale Experimental Procedure
The following is a summary of the procedures for the laboratory scale experiments.
The soil is mixed with enough water to give a 3:1 water to soil weight
ratio, and then mechanically stirred so there is no accumulation in the
container.
A pH controller is used to bring the pH to the setpoint. The amount of acid
used is recorded, and the extraction is continued for the desired amount of
time.
The mixer is shut off and the soil is allowed to settle.
After the soil settles (approximately 30 minutes), the extractant is
decanted. Deionized water is added to rinse the soil, and this is also
removed after the soil settles again.
Samples of the soil are removed for total metals and TCLP analysis. A
second extraction may be run on the soil if desired.
1.4.2 Laboratory Scale Results
The bulk of the first-year experiments were performed on the Synthetic Analytical
Reference Material (SARM) and the Synthetic Soil Matrix (SSM), both from the EPA,
as well as soil from a Superfund site in Pennsylvania. Table 1 below shows the
reduction of TCLP metals concentrations in the SARM soil during a typical two-step
extraction at a pH of 2.
Table 1 SARM Laboratory Scale Extraction
Metal
As
Cd
Cu
Ni
Pb
Zn
TCLP (mg/L)
Untreated
Sand
5.5
20
140
17
36
650
After One '
Extraction
2
2.4
3.6
0.5
14
9.6
After Two
Extractions
1.4
1.5
8.3
0.9
46
18
Percent Reduction
One
Extraction
64%
88%
97%
97%
61%
99%
Two
Extractions
75%
93%
94%
95%
N/A
97%
-------
Parametric experiments were used to determine the optimal pH, residence times and
number of extraction steps. The best results were achieved using a one- or two-step
process with 15 to 30 minute residence times, The following conclusions were drawn
about the Acid Extraction Treatment System:
The AETS process using hydrochloric acid is a viable means of reducing
TCLP metals to levels below the RCRA limits for soils contaminated with
As, Cd, Cu, Ni, Pb, and Zn.
The process can reduce the soil's total metal concentrations by 60 to 99
percent.
Although a single extraction step is sufficient in many cases, the system
should be designed with sufficient flexibility to accommodate two steps.
The second year of the investigation was planned based on the results of the
laboratory scale experiments.
-------
2.0 SUMMARY OF SECOND YEAR ACTIVITIES
The following table is a brief chronological history of the events that took place
during the second year of the AETS project.
EVENT
Beginning of second year.
Ideas for the preliminary system designs were being pulled
together, as well as some minor equipment purchases, and the
mass balance calculations were started.
Decisions were made to use a cyclone separation system
rather than a screw-type conveyor system. More soils were
needed and soils should be characterized for particle size
distribution, total metals, and TCLP.
Preliminary system designs were drawn up. More soil was
acquired from the EPA (SSM), and a Superfund site in
Palmerton, PA. The SSM soil contained As, Cd, Cr, Cu, Ni,
Pb, and Zn. The Palmerton soil is mainly contaminated with
Cd, Cu, Pb, and Zn.
The tanks, mixers, piping, and 1-inch and 2-inch cyclones
needed to construct the pilot-scale acid extraction system
were purchased.
Soil from a Superfund site in Pedricktown, NJ was acquired.
This soil was contaminated with high concentrations (>3%) of
lead and lower concentrations of copper and zinc. The
attrition scrubber, pH controller, and rubberized centrifugal
pump were purchased.
The acid regeneration system was acquired.
The cyclones and acid regeneration system were installed and
tested.
The attrition scrubber, up-flow filter, rubber pump, and pH
controller were installed and tested.
Soil from a Superfund site in Butte, Montana was acquired.
This soil is contaminated with copper and zinc. The initial
experiments (AE-102 through AE-108) were performed using
the pilot-scale system seen in Figure 2.
DATE
4/91
6/91 - 8/91
10/91
2/92
2/92-3/92
4/92
5/92
6/92
7/92
8/92
-------
Soil from a site in King of Prussia, NJ was acquired. The main
contaminants were copper, chromium and nickel. The
rubberized pump was rendered unusable due to the abrasion of
the soil.
10/92
A new slurry pump was ordered, installed, and tested. The
new pump has an acid resistant impeller and inner coating, and
can pump particles up to l-inch in diameter, which eliminates
the abrasion problems.
10/92-
11/92
The second set of extractions, AE-112 to AE-117, was
performed with only slight procedure changes from the first
set, including the new slurry pump.
11/92
The entire process flow was changed to include a second
pump and a second cyclone. The final set of extractions (AE-
118 through AE-122) was run using this new configuration,
which can be seen in Figures 3 through 5.
12/92
Follow-up experiments on the clay fractions of soils were
performed. Agricultural tests were performed on treated soils.
1/93 - 2/93
2 .1 SUMMARY OF EXPERIMENTAL PROCEDURES
The following experiments were performed on the pilot-scale acid extraction unit
designed and built by CHMR. The experimental procedures are separated into three
groups (the experiments are numbered in the text in the same format they were
numbered in the laboratory):
The first set of extractions, AE-102 through AE-108, which featured a static
screening process for the initial soil classification.
The second set of extractions, AE-112 through AE-117, which used soils were
screened with an automatic shaker and a slurry pump.
The third and final set of extractions, AE-118 through AE-122, which utilized
a modified pilot-scale system.
The procedures used for the three sets of extractions are discussed in more detail in
subsequent sections.
-------
TO ACCUMULATION
AT EXPERIMENT'S END,1
FILTER/
AGIO
REGENERATION
PUMP
*THE SOIL IS ACCUMULATED TO BE RINSED, AND
THE EXTRACT ANT IS ACCUMULATED FOR FINAL REGENERATION.
Figure 2 Extraction Flow Diagram
2.1.1
First Extraction Procedure
Initial Preparations
Sieve soil to below 9 mesh after drying it overnight at 80°C.
Test pumps, cyclones, valves and other parts of the AETS system to
ensure that they are working properly before beginning the experiment.
Refer to Figure 2 for the process flow diagram,
8
-------
Put the soil in the attrition scrubber at a ratio of approximately 60
percent solids by weight and scrub for the allotted time.
Rinse the scrubber with enough liquid (either fresh water or previously
regenerated extractant) to get a ratio of 20 percent solids by weight in
the extraction vessel.
Begin the extraction by turning on: the mixer, pump to the 2-inch
cyclone, and the pH controller.
At the desired pH, the cyclone overflow is routed to the acid
regeneration system, and then back to the extraction tank. The
underflow is sent directly to the extraction tank. Continue the extraction
for the desired time.
At the end of the extraction, route the cyclone underflow to a 100 mesh
screen for dewatering, and route the overflow to storage tanks for later
regeneration.
Regenerate the acid and rinse the soil in the manner specified.
The filter size used was 1 micron.
Repeat extraction procedure for the second extraction.
The equipment used for these extractions is listed in Appendix A.
2.1.2 Procedure Changes for Second Set of Extractions
The extracted soil in the cyclone underflow is dewatered using a 200-
mesh mechanical shaker, which allows for faster separation.
Single step extractions are performed at a pH of 2.0 for 40 minutes,
with soil samples being taken every 10 minutes, to determine the effects
of residence time distribution. Timed soil samples are rinsed in a beaker
to simulate tank rinsing.
The extracted soil is rinsed using a few different methods, i.e. in the
tank or on the shaker, depending upon the experiment.
A 5 micron cartridge filter is used to separate clay from the extractant.
2.1.3 Procedure Changes for Final Extractions
The plumbing was changed to add another pump and cyclone, and a
settling tank to reduce the amount of clay sent to the filter and acid
regeneration system.
-------
Extractions for this set are single step extractions at a pH of 2.0 for 20
minutes.
The new process flow diagrams for the extraction and dewatering steps
can be seen in Figures 3 and 4. The underflow from the 1-inch cyclone
shown in Figure 4 is put back on the 200-mesh screen to trap the fines
in the sandy-soil fraction.
Figure 3 Revised AETS Flow Diagram
A standard rinsing procedure was developed for the final set of
extractions.
The soil is placed in the rinse tank with the appropriate amount of
rinsate.
The soil is mixed for the desired amount of time and then drained
onto the shaker, where the solids are separated, and the rinse is
collected for regeneration.
The rinsate is regenerated prior to being reused in the next
experiment, and the soil is the finished, clean, soil. See Figure 5
for the rinsing flow diagram.
10
-------
SYSTEM IS RUN UNTIL MOONO TANK IS EMPTY
Figure 4 Dewatering Flow Diagram
The 1-inch cyclone underflow in Figure 5 is put on the shaker to
"trap" the fines in the sandy-soil fraction. The overflow is used
to rinse all of the soil out of the tank, due to the batch process
being used.
2.1.4
Laboratory Analytical Procedures
The samples were analyzed according to the analytical procedures outlined in the
Quality Assurance Project Plan (QAPP). These included the standard EPA method for
TCLP extractions (SW-846 1311), and SW-846 6010 for total cadmium, chromium,
copper, nickel, lead, and zinc analyses. Method SW-846 7061 was employed for
arsenic analyses.
The Quality Assurance/Quality Control procedures employed are summarized in
Appendix
11
-------
KEY;
-RfOFtCUtATHJ UNTIL
NO SOUOS BgMAIN i
W THE RN9NG TANK i
- i=»OU7I2 TO ACC I
REGENERATION WHEN AU.'
SOUOS We S6MOVS3
f BOM M;XINQ TANK
SOLOS
Figure 5 Rinsing Flow Diagram
2.2 SUMMARY OF PILOT-SCALE EXPERIMENTS
A total of 18 extractions were performed. The following table gives a description of
the extractions, with a description of procedure adjustments made to each specific
experiment.
EXPERIMENT
AE-102
AE-103
AE-104
AE-105
AE-106
SOIL USED
Butte
Butte
Butte
Synthetic Soil Matrix
. Synthetic Soil Matrix
CONDITIONS/COMMENTS
No concurrent regeneration
Pump malfunction/exper. stopped
Regeneration during extraction;
rinsing on screen
Fines (< 50 microns) removed prior
to extraction
Fines removed prior to extraction
12
-------
AE-107
AE-108
AE-112
AE-113
AE-114
AE-115
AE-116
AE-117
AE-118
AE-119
AE- 120
AE-121
AE-122
Pedricktown
Palmerton
Butte
Butte
Pedricktown
King of Prussia
King of Prussia
King of Prussia
King of Prussia
Butte
Synthetic Soil Matrix
Synthetic Soil Matrix
King of Prussia
Soil rinsed on screen and in tank
Extraction stopped due to pump
malfunction; usable data obtained,
though
Single step extraction
Samples taken over range of
residence times to assess process
Extraction tank plumbing altered
prior to extraction; new pump used
pH meter malfunctioned, no usable
data obtained
Cyclone required adjustment
Mass balance performed
10 micron filters used
13
-------
3.0 RESULTS AND DISCUSSION
In this section, the experiments are categorized based on the types of soil used. The
experimental data, soil description, particle size distribution, and discussion will be
presented for each type of soil. The following is a list of the soils used, and the
number of successful extractions performed on each one with the pilot-scale system.
Butte, Montana soil - 4 extractions
King of Prussia, NJ soil - 4 extractions
Synthetic Soil Matrix (SSM) - 4 extractions
Pedricktown, NJ soil - 2 extractions
Palmerton, PA soil - 1 extraction
The results from these extractions are discussed in subsequent sections.
3.1 TREATMENT GOALS
The treatment goals for metal-contaminated soils include two criteria -- Toxicity
Characteristic Leaching Procedure (TCLP) and total metals.
Table 2 Total and TCLP Metal Treatment Requirements
i California
Metal
As
Ag
Ba
Cd
: Cflll
: CrV!
' Cu
: Hg
; Ni
; Pb
Total fnntj/kg}_
500 i
500 i
10,000 :
100 |
2,500 ;
500
2,500 \
20 j
2,000 '
1 ,000 ':
' Se 100 i
Zn ; s.ooo x
TCLP (mcj/
5,00
5,00
100.00
1.00
560.00
5,00
2.5,00
0,20
20,00
5,00
1.00
250.00
i U. S. EPA
Ll- Total {E|9/kJLTCLliirrLa/L
; 5.00
: o.so
; '. 100.00
i 1.00
; i 5.00
j I
0,20
! 500-2000** i 5,00
: I 1.00
j j
K.O.P. Site*
)_Total (rng/kg)
i 483
; 3,571
: 1,935
Treatment objectives for Heidemij Reststoffendiensten Soil Washing Demonstration.
**500 mg/kg in surface soil, 2000 mg/kg below the surface.
TCLP is a measure of the leachability of metals in soil. Salient TCLP maximum
concentrations are set in RCRA for 7 metals. The limits are shown in the third column
of Table 2. TCLP levels for additional metals have been set by the state of California.
These, also are given in Table 2.
14
-------
Total metals concentration limits are not set by Federal statute. They are often set
by individual EPA Project Officers for specific sites. Sample total metals limits are
given for the King of Prussia Superfund site. In addition, the EPA has set an action
level of 500 mg/kg for lead in surface soils, and 2,000 mg/kg for lead in soils 2 feet
or more below the surface. California has set total metals limits for a variety of
metals. These limits are summarized in Table 2.
3.2 BUTTE, MONTANA SOIL
Site History
The Superfund site in Butte was placed on the National Priorities List (NPL) because
of a potential dust hazard on the site. The soil is non-hazardous, because the main
contaminants are copper and zinc, both of which are not considered by RCRA to be
hazardous metals. The Butte soil was specifically chosen because it was non-
hazardous, and any residuals produced during experiments would also be non-
hazardous.
Soil Description
The screening results and particle size distribution are given in Table 3 and Figure 6.
The figures show that 2 percent of the soil is smaller than 20 microns, and 6 percent
is smaller than 60 microns. The initial soil had the texture of beach sand that had
very little clay and large (>1/8") particles. The dried soil had a density of 1.25 g/cc,
and the average total copper and zinc levels are 1250 mg/L and 116 mg/L,
respectively.
Table 3 Butte, Montana, Screen Analysis
Mesh
+5
+9
+20
+40
Micron
4000
2190
841
420
+60 250
+100 149
-100 N/A
Volume(ml)l % on
100 on
3.9
100 on 3.9
250 on
575 on
750 on
450 on
325 thru
9.8
22.5
29.4
17.6
12.7
The soil is yellowish in color. Examination under a microscope revealed small
turquoise and orange particles, some apparently attached to the surface of sand
particles, and some scattered throughout the sand. There were no apparent size
differences among the sand particles with or without metals on the surface.
15
-------
^ 40 3
| O
2«8 3K 488 704
Figure 6 Butte Soil Particle Size Distribution
Discussion of Results
Seven experiments (AE-102, 103, 104, 112, 113 and 119) were attempted using the
Butte soil. One experiment (AE-103) was cancelled due to a pump malfunction, and
no data for AE-113 were obtained because the samples were destroyed during
shipping. AE-102 and AE-104 were duplicate experiments run under identical
conditions, with a two-step extractions and 20 minute residence times. AE-112 was
a single stage extraction. This experiment was run to see how reproducible the
removal levels were with new equipment. There were two main objectives for AE-
119, the final Butte extraction. The first was to recover all of the soil to form a mass
balance, and the second was to analyze the different soil fractions, i.e., the filter
sludge, the sandy soil.
The data in Tables 4, 5, 6, and 7 show the results from experiments successfully
conducted with the Butte soil. The tables show the TCLP concentrations for the
untreated and treated soils (either after the first or second extractions), and the total
metals concentrations in the untreated and treated soils. The soils were analyzed
primarily for copper and zinc, with some analyses for lead and arsenic as well. The
data show that the acid extraction system worked well for this particular soil. The
initial TCLP and totals concentrations for the Butte soil were within the federal and
California requirements.
16
-------
Table 4 AE-102: Butte Soil, 1st Experiment
Metal
Zn
Cu
TCLP (mg/L)
Untreated
2.6
1.4
1st Ext
N/A
N/A
2nd Ext
0.0
0.1
% Removal
1st Ext | 2nd Ext
N/A 100%
N/A 91%
Metal
, Zn
Cu
Total Metal (mg/kg)
Untreated
1350
114
1st Ext | 2nd Ext
499 ! 285
31 201
% Removal
1st Ext ! 2nd Ext
63% j 79%
73% 82%
Table 5 AE-104: Butte Soil, 3rd Experiment
TCLP (mg/L) 1% Removal
Metals
Cu
Zn
J Untreated I
i 1.71
! 7.1!
2nd Ext I
0.05i
0.71 i
2nd Ext
97%
90%
i Metal
| As
j Pb
! Cu
! Zn
Total Metalsjmc
Untreated! 1st Ext
1 83 j 52
390 I
127J 26
1250] 376
ji/ sLt
2nd Ext !
64 i
49!
21 |
i_ 326 j
% Removal
1st Ext ! 2nd
72% i
i
80% i
70% i
Ext :
65% ,
87% !
83% '
74% '
Table 6 AE-112: Butte Soil, 4th Experiment
Metal
TCLP
Initial
Final
% Removal;
Cu
Zn
2.7
0.2
1,2
86% :
56% '
Metal
Cu
Zn
Total Metals (mg/kg)
Initial
98.0
1,170.0
Final
18.0
195.0
% Removal
82%
83%
17
-------
Table 7 AE-119: Butte Soil, 6th Experiment
I Total Metals (mg/kg) |
!
i
1
I
i
Metal !
Cu !
Pb j
Zn .j
Initial ! Final j% Removal!
125|
338 j
1,230!
27! 78% |
39 88%
118| 90% |
TCLP concentrations for copper and zinc were reduced between 55 and 100 percent
of the initial levels. The total metals were reduced an average of 82 percent. In the
cases where lead levels were analyzed, the system removed an average of 88 percent
of the lead. The results indicated modest improvements in system efficiency using
the better rinsing and dewatering procedures of the later experiments. The results
demonstrated that AETS is capable of removing metals, even starting at relatively low
concentrations.
Although the Butte soil contains a small amount of clay, no separate clay removal and
disposal is anticipated to be necessary. Approximately 2 percent of the soil was
retained on the filter prior to the acid regeneration system. This soil was analyzed for
copper, lead and zinc and found to contain 380, 3500 and 980 mg/kg of each,
respectively. The filter sludge could be mixed with the remaining soil. With the filter
sludge mixed in, the composite result would have been a 75 to 90 percent reduction
in metals concentration.
The significance of the Butte soil was its usefulness in determining experimental
parameters using a non-hazardous soil, and a demonstration that metals can be
removed efficiently even starting in a relatively low concentration range.
3.3 KING OF PRUSSIA, NJ SOIL
Site History
The King of Prussia Superfund site is located in Winslow Township, New Jersey. The
facility on the site was used to neutralize acid streams from another facility located
adjacent to the site. The contaminants of interest in this soil are chromium, copper,
and nickel. The material is not RCRA hazardous. The site is listed on the Superfund
NPL because of the high levels of chromium. In addition, the soil at the site was
recently used to demonstrate another soil washing system (which works primarily by
particle size separation). By incorporating the soil into AETS testing program, direct
comparisons between the two systems are possible.
Soil Description
The screening results and particle size distribution are given in Table 8 and Figure 7.
Seven percent of the soil is smaller than 20 microns, and 11 percent is smaller than
18
-------
100 microns. This soil was light tan in color, with visible yellow clays and some large
particles. The dried soil has a density of 1.34 glee.
Table 8 King of Prussia, NJ, Screen Analysis
Mesh
+20
+40
-40
Micron
841
420
N/A
Weight (g)
295 on
81 on
127 thru
% on (thru)
59
16
(25)
100
Microni
Figure 7 King of Prussia Particle Size Distribution
Microscopic examination revealed black, turquoise, and orange-red particles. The bulk
of the soil is orange-red. The black and turquoise particles may be pieces of
contamination.
Discussion of Results
Tables 9, 10, 11, and 12 summarize the results from experiments performed using
this soil. These tables show the initial and final metals concentrations in the soils.
Experiments AE-115 and AE-116 were intended to study the effects of residence time
distribution. Soil samples were taken every ten minutes and rinsed by simulating tank
19
-------
Table 9 AE-115: King of Prussia Soil, 1st Experiment
Metal Removal Efficiency from the Soil
Metal
Cr
Cu
Ni
Total Metals (mg/kg)
Initial
1020
1240
335
Final
37
17
4
% Removal
96%
99%
99%
Timed soil samples
Soil
10 minute
20 minute
30 minute
40 minute
Total Metals (mg
Cr (mg/kg)
135.0
78.7
51.9
73.7
Cu (mg/kg)
33.1
18.1
13.3
21 .0
/kg)
Ni (mg/kg)
12.1
5.1
3.2
4.9
Table 10 AE-116: King of Prussia Soil, 2nd Experiment
Metal Removal Efficiency from the Soil
Metal
Cr
Cu
Ni
Total Metals (mg/kg)
Initial soil
1240
1660
518
Final soil
89
18
6.3
% Removal
93%
99%
99%
Timed soil samples
Soil
10 minute
20 minute
30 minute,
Total Metals (mg/kg)
Cr
224
183
130
Cu
47
40
30
Ni
16
15
8
20
-------
Table 11 AE-118: King of Prussia Soil, 3rd Experiment
Removal Efficiency from Soil
Metal
Cr
Cu
Ni
Total Metals (mg/kg)
Initial*
205
293
50
Final
45
47
11
% Removal
78%
84%
79%
These results are inconsistent with other samples.
Table 12 AE-122: King of Prussia Soil, 4th Experiment
Metal Removal Efficiencies from the Soil
Metal
Cr
Cu
Ni
TCLP (mg/L)
Initial
0.2
7.1
27.6
Final
0.1
0.2
1.2
% Removal
59%
97%
96%
Metal
Cr
Cu
Ni
Total Metals (mg/kg) [
Initial
1390
2030
514
Final
324
93
12
% Removal
77%
95%
98%
Other Soil Samples
Sample & Description
122-S-07 Rinsed settled soil
Rinsed settled soil TCLP
122-S-08 Rinsed filter sludge
Rinsed filter sludge TCLF
TCLP in mg/L; metals in mg/kg
Cr
4780
0.8
3390
0.3
Cu
707
0.4
760
1.1
Ni
97'
0.1
100
2.8
21
-------
rinse conditions in a beaker. Results of analyses of these samples are also presented
in Tables 9 through 12. TCLP analyses were not performed because TCLP was not a
concern with the site. The residence time data show significant decreases in total metals
concentration between 10 and 20 minutes, with the metals concentrations still decreasing,
but flattening after 20 minutes.
Differences between the 40-minute "timed" data and the final soils when 40-minute
extractions were used may be accounted for by differences in the rinse methods used.
The bulk of the soil was better rinsed than the small samples removed throughout the
extractions.
AE-115 and AE-116 were identical experiments. The concentrations in AE-116 are
slightly higher, and this is because the soil in AE-116 was rinsed with regenerated
rinsate from AE-115. All of the metals may not have been removed from this rinsate.
AE-118 was the first extraction using the new process flow in Figures 3 through 5.
AE-118 was performed under conditions similar to AE-115 and AE-116, except a 20-
minute residence time was used. The initial soil analytical results 'appear to be in
error, since they were about 80 percent lower than other experiments using the same
soils. The results from AE-118 indicate that copper and nickel removal was
approximately the same for either 20 or 40 minute residence times. However,
chromium was removed more efficiently when longer extractions were performed.
The objectives for AE-122 were to develop a soil balance and analyze the different
fractions of the soil, in order to determine if all fractions of the soil could be
recombined to produce a final soil which met the treatment criteria given in Table 2.
The results are summarized in Table 13. The coarse solids, which are the bulk of the
soil (80 percent or more), were well within all prescribed treatment standards. The
fines fraction showed good removal efficiency for Cu and Ni, and was within the limits
for these metals. The fines were not within the treatment limit for Cr. It appears that
only about half the chromium was actually removed from the soil.
Table 13 Overall King of Prussia Results with Comparison to Heidimij Results
I
Metal
Cr
Cu
Ni
Treatment
Standard
(mg/kg)
463
3571
1935
Heidimij
Result
(mg/kg)
170
350
70
AETS Results
Avg Cone.
Final Sand
(mg/kg)
123
44
8
Avg Cone.
in Fines
(mg/kg)
2705
646
100
Composite
Result
(mg/kg)
639
164
26
Overall
Removal
Efficiency
47%
90%
94%
Based on average concentrations computed over all extractions performed, if the fines
were mixed back in with the soil, the resulting composite would not meet the site
treatment standard of 483 mg/kg chromium. However, the composite values
22
-------
represented a combination of 20 and 40 minute extraction results. Based on data
exclusively from AE-115 and AE-116, which were performed with longer residence
times, 40-minute extractions would remove additional chromium, producing a
composite with 470 mg/kg chromium, which is within the treatment standard.
Overall, the King of Prussia soil - maintained as a whole - appears to be treatable
with the Acid Extraction treatment System, using a 40-minute residence time.
3.4 SYNTHETIC SOIL MATRIX
Soil History
The Synthetic Soil Matrix (SSM) was developed by the EPA specifically for use in
research and development of emerging or innovative technologies. It is a mixture of
clay, sand, silt, gravel, and topsoil that is blended together to form the soil matrix.
Organic and inorganic contaminants are added based on typical hazardous materials
at Superfund sites. These experiments used SSM soil which contained high levels of
metals and no organic contaminants. The metals in the SSM were arsenic, cadmium,
chromium, copper, lead, nickel, and zinc.
Soil Description
The screening results and particle size distribution are given in Table 14 and Figure 8,
respectively. Approximately 30 percent of the soil is under 50 microns. This soil has
a significant amount of clay, and few large particles. The density of the soil was
found to range between 1.29 and 1.34 glee.
Table 14 Synthetic Soil Matrix Screen Analysis
Mesh
+9
+20
+40
+100
-100
Micron
2190
841
420
170
N/A
Weight % on
4
16
15
8
57
According to its specifications, 40% of the SSM soil is
below 75 microns. 12% is below 5 microns.
Microscopic analysis of the soil revealed only a few small particles of metals. The
bulk of the particles are medium brown in color, with a few black flecks throughout.
In addition, orange and turquoise colored particles were visible. Some of the sand
particles appeared to be coated with the turquoise metals.
23
-------
100
s -
1* 1.9 2.8 J» 55 7.8
16 22 31 *» 62 M 12* 17S 2*9 352 498 70*
Micron*
Figure 8 Synthetic Soil Matrix Particle Size Distribution
Discussion of Results
Four extractions were performed on Synthetic Soil Matrix. The results are presented
in Tables 15, 16, 17, and 18, which give the untreated and treated soil TCLP and
total metals values, as well as the reduction/removal efficiencies for the process.
Table 18 also shows the analytical results from other process samples taken. Total
metal removal efficiencies typically ranged between 80 and 90 percent for cadmium,
copper, chromium, nickel, lead, and zinc. The arsenic removals only averaged 43
percent for total metals, however. This is consistent with the results of the first-year
pH studies which showed that the best arsenic removal occurs at a pH of 1.
The system showed significant TCLP reductions. With a few exceptions, the percent
reduction for copper, nickel, and zinc (not RCRA hazardous) was in the 90 percent
range. In every extraction, the TCLP for lead was successfully reduced below the
RCRA limit from initial levels as high as 27 mg/L. Chromium and arsenic TCLP
concentrations were maintained below RCRA limits. The percent reduction of TCLP
arsenic significantly improved in the final two SSM extractions, and the zinc reduction
was slightly lower. TCLP concentrations for cadmium remained above the RCRA
limits.
Experiments AE-105 and AE-106 were performed under identical conditions to see if
acid extraction would work on this soil, and to determine how repeatable the results
were. AE-105 and AE-106 were performed using only the coarse fractions of the
24
-------
Table 15 AE-105: Synthetic Soil Matrix, 1st Experiment
Metal Removal Efficiency from the Soil
1
Metal | Untreated
As
Cd
Cu
Cr
Ni
Pb
Zn
Metal
As
Cd
Cu
Cr
Ni
Pb
Zn
3.8
8.3
130.0
0.1
11.8
6.9
517.0
(mq/l_;
1st Ext
2.4
2.2
15.2
0.1
2.2
2.3
154.0
2nd Ext
2.6
1.4
4.8
0.1
0.7
0.6
45.2
% Removal
1st Ext
38%
74%
88%
NA
81%
67%
70%
2nd Ext
33%
84%
96%
NA
94%
92%
91%
Total I Metal (m a/kg)
Untreated
180'
254
5340
451
434
1820'
14000
1st Ext
450
141
2nd Ext
155
53
472 213
30 16
78 29
185 112
3560 1190
% Removal
1st Ext
NA
44%
91%
93%
82%
90%
75%
2nd Ext
14%
79%
96%
96%
93%
94%
92%
Note: Clays were removed from 'untreated' soils for AE-105
prior to the experiment.
Table 16 AE-106: Synthetic Soil Matrix, 2nd Experiment
Metal Removal Efficiencies from the Soil
Metals
Metal
As
Cd
Cu
Cr
Ni
Pb
Zn
As
Cd
TCLP .(mg/L)
Untreated | 2nd Ext
0.6 2.1
8.4 1.6
Cu 106 3.4
Cr 0.2 0.1
Ni
Pb
Zn
Total.
Untreated
304
246
2870
195
254
985
9520
11.6
24.4
501
0.6
1.2
26.2
% Removal
2nd Ext
NA
81%
97%
65%
95%
95%
95%
Metals (mg/kg)
1st Ext
225
80
359
24
36
370
1305
2nd Ext
111
57
125
17
19
196
599
% Re
1st Ext
26%
67%
87%
88%
86%
62%
86%
nova I
2nd Ext
63%
77%
96%
91%
93%
80%
94%
Note: Clays were removed from 'untreated' soils for
AE-106 prior to the experiment.
25
-------
Table 17 AE-120: Synthetic Soil Matrix, 3rd Experiment
Metal Removal Efficiencies from the Soil
Metal
As
Cd
Cr
Cu
Ni
Pb
Zn
TCLP (mg/L)
Initial
4.0
41.0
<0.05
297.0
35.6
27.1
669.0
Final
0.8
2.1
0.1
21.1
2.8
2.8
220.0
% Remd
81%
95%
N/A
93%
92%
90%
67%
Metal
As
Cd
Cr
Cu
Ni
Pb
Zn
Total Metals (mg/kg)
Initial..
620
970
1,320
10,900
979
10,040
20,500
Final J % Removal
530
95
63
536
70
213
3.080
15%
90%
95%
95%
93%
98%
85%
The initial concentrations and TCLP include the fine fractions of the soil.
Other Solid Samples
Sample & Description % of Soil
Rinsed settled soil Total 12%
Rinsed settled soil TCLP
Estimated Initial Fines Cone.
Rinsed filter sludge Total 22%
Rinsed filter sludge TCLP
mg/L for TCLP metals, mg/kg for soil Sample metals
As
450
0.7
630
1070
1.1
Cd
33
1.0
686
25
0.8
Cr
250
0.8
1980
1430
1.1
Cu
837
21.5
17200
1350
19.0
Pb
2730
81.2
19500
31700
101.0
Ni
58
15
1370
82
1.5
Zn
1800
44.3
25200
1090
21.1
Table 18 AE-121: Synthetic Soil Matrix, 4th Experiment
Metal Removal Efficienct from the Soil
Metal
As
Cd
Cr
Cu
Ni
Pb
Zn
TCLP (mg/L)
Initial
4.2
48.9
<0.05
298.0
35.9
26.0
719.0
Final
0.7
2.4
<0.05
12.1
2.8
2.3
154.0
% Remova
84
95%
N/A
96%
92%
91%
79%
Metal
As
Cd
Cr
Cu
Ni
Pb
Zn
Total Metals (mg/kg)
Initial
730
1,130
1,640
12,400
1.410
10,800
26,300
Final % Removal
150
188
93
641
128
315
4,760
79%
83%
94%
95%
91%
97%
82%
The initial concentrations and TCLP include
the fine fractions of the soil.
Other Solid Samples
mg/L for TCLP metals,, mg/kg for soil sample metals
Sample & Description % of Soil
Rinsed settled soil Total 10%
Rinsed settled soil TCLP 10%
Estimated Initial Fines Cone.
Rinsed filter sludge Total 33%
Rinsed filter sludge TCLP 33%_
As Cd |Cr Cu Pb Ni Zn
170
0.6
630
670
1.2
20
0.5
686
35
1.2
153
0.4
1,980
727
0.7
346 1,200 27 745
7.9 26.2 0.7 21.0
17,200 19,500 1,370 25,200
1,106 24,070 86 1.170
20.3 50.8 2.1 32.5 '
26
-------
soil - the fines were removed using a hydrocyclone before the extraction. The total
metals concentrations during these extractions were lower than those for the final two
SSM extractions.
AE-120 and AE-121 were performed using the new system configuration with the
same objectives as before: complete soil recovery and analysis. The TCLP for the
two fine soil fractions (filter sludge and settled soil, which represent approximately 40
percent of the original soil by weight) was within treatment standards, except for lead
and one cadmium value. The total metal concentration in the fines was within the
salient California or EPA standards for all metals except for lead and arsenic.
Based on the quantities of soil recovered in each fraction and the total metal and TCLP
concentrations in each recovered fraction, a composite "final" soil concentration was
calculated. These composite results are given in Table 19. The results show that the
composite total metals concentrations for arsenic, cadmium, chromium, copper,
nickel, and zinc were within the treatment objectives. The only metal which was not
extracted in appreciable quantities was lead. The lead appears to be concentrated in
the filter sludge, which is the finest soil fraction.
Table 19 SSM Soil Composite Results Using Data from AE-120 and AE-121
~Me"taT
As
Cd
Cr
Cu
Ni
Pb
Zn
Cone, in
Coarse (60%)
75
94
47
321
64
158
2380
Cone, in
Fines (40%)
280
20
300
500
37
9500
540
Composite j TCLP in ;
Cone, I Coarse '
160 0,3;
60 1.2
150 0.0 ;
390 6,1 ;
50 1.4i
3890 1.2;
1640; 77
TCLP in
Fines
0.5
0.5
0.3
9
0.9
23
15,0
Composite
TCLP
_
0,4
0.9
1.2;
10:
50 !
The composite samples show compliance with TCLP limits for everything but lead and
cadmium. Given its apparent extractability, the cadmium may be removed by using
a longer rinse or more rinsewater.
The TCLP and total metals concentration for lead would have been within the EPA and
California regulatory limits if the coarse sand and rinsed settled soil (the coarser fines
from the cyclone overflow) were composited together. The remaining fines, which
represent between 20 and 25 percent of the original soil, would have to either
undergo additional treatment for the removal of lead or stabilization and disposal.
Overall, the extractions using SSM showed that five of the six metals (all except lead)
in the soil are readily treated using 20-minute residence times to the prescribed limits
for total metals, and four of the six metals (all except cadmium and lead) are treatable
to the TCLP limits. Based on the first year and results from other soils, the cadmium
27
-------
may be treatable using a 30 to 40-minute extraction, coupled with more extensive
rinsing. Lead is treatable in the bulk of the soil provided that the fines (representing
20 to 25 percent of the SSM) are first removed. These fines may be amenable to
treatment using different methods.
3.5 PEDRICKTOWN, NJ SOIL
Site History
Soil was obtained from the National Lead Superfund site in Pedricktown, NJ. The
facility crushed and processed lead-acid batteries through an on site furnace for lead
reclamation. The soil was contaminated with copper, lead, and zinc, but was selected
for testing primarily due to its high levels of lead.
Soil Description
The screening results and particle size distribution are given in Table 20 and Figure 9,
respectively. The soil is sandy, with some clays and few large particles (> 1/5")
present. The raw soil is reddish in color with some visible clays. Eight percent of the
soil would pass through a 150 micron screen. The density of the soil ranged from
1.58 to 1.89 g/cc. The average initial TCLP lead is 510 mg/L, and the average initial
total lead is 26,200 mg/kg, which are high, even compared to the SSM. Table 20
shows the distribution of lead throughout the different fractions. The concentration
of lead in the fines (-100 mesh) was extremely high (over 13 percent), but even the
coarse fractions of the soil contained appreciable quantities of lead. The bulk of the
lead (over 70 percent) was present in the fraction of soil between 150 and 850
microns in size.
Table 20 Pedricktown, NJ, Screen Analysis
Mesh
+5
+9
-1-20
+40
+60
+ 100
-100
Micron I Weight (g)
4000 !! 124 on
2190
841
420
250
149
N/A
i 160 on
342 on
550 on
458 on
216 on
160 thru
% on
6.2%
8.0%
17.1%
27.5%
22.9%
10.8%
Pb (mg/kg]
12,000
12,000
34,500
34,500
34,500
34,500
8.0% 132,000
Overall: 33,000
Microscopic examination revealed that the soil was very grainy, with orange-red
particles throughout the granules. There were also many very dark particles mixed in
with the soil. The bulk of the particles appeared to be beige or yellowish-orange in
color.
28
-------
I
Q
4S8 70*1
Figure 9 Pedricktown, NJ, Soil Particle Size Distribution
Discussion of Results
Two extractions were performed using the Pedricktown soil, AE-107 and AE-114.
The results from these extractions are presented in Tables 21 and 22, which show the
initial and final TCLP and metals concentrations. Data from samples taken at different
residence times are also given in Table 22. The AETS system dramatically reduced
the TCLP lead in both extractions. The TCLP concentration was reduced by 95
percent in one experiment and by 99 percent in the other. During the first
experiment, the concentrations were reduced to just above the RCRA limit of 5 ppm.
Based on these results, CHMR anticipates that the TCLP can be reduced below the
RCRA limit if a longer residence time is used. The rinse system for the treated soil
malfunctioned during the second extraction. Inefficient rinsing tends to increase the
TCLP values for the soil, because the entrained acid in the soil contains high
concentrations of highly mobile metal contaminants. The high TCLP values for this
soil were attributed to the problems with rinsing. However, a confirming experiment
could not be performed because CHMR had insufficient soil.
The data for AE-107 shows that a 20 minute, one-step extraction can remove the
majority of the lead from the Pedricktown soil. The residence time data from
experiment AE-114 show that more than 90 percent of the lead was removed after
only five minutes. The final total lead concentration is above the EPA surface soil
limit, but well below the EPA limit for soils two feet below the surface. The total level
is approximately equal to the California treatability limit for lead.
29
-------
Table 21 AE-107: Pedricktown Soil, 1st Experiment
Metal Removal Efficiencies from the Soil
Metal
Pb
TCLP (mg/L)
Untreated 2nd Ext
520.0 | 5.1
% Removal
2nd Ext
99.02%
Metal
Pb
Cu
Zn
Total Metals (mg/kg)
Untreated
29,200.0
192.0
239.0
1st Ext
1,430.0
92.3
345.0
2nd Ext
1,310.0
68.3
195.0
% Removal
1st Ext
95.10%
51.93%
NA
2nd Ext
95.51%
64.43%
18.41%
NOTE: 1st extraction soil was dewatered and rinsed on the
screen. The 2nd extraction soil remained in tank overnight, then
rinsed on the screen.
Table 22 AE-114: Pedricktown Soil, 2nd Experiment
Metal Removal Efficiency from the Soil
Metal
Pb
TCLP (mg/L)
Initial
503.0
| Final
23.1
% Removal
95.41%
Metal
Pb
Total Metals (mg
Initial
23200.0
Final
1,040.0
/kg)
% Removal
95.52%
Timed soil samples
Soil
5 minute
1 0 minute
20 minute
30 minute
40 minute
Pb(mg/kg)
1,790
1,930
2,210
954
1.080
The purpose of AE-114 was to find a suitable residence time for this soil. Although
95 percent of the total lead is removed after 5 minutes, some more studies may be
needed to determine how long is necessary to reduce the TCLP lead to below
acceptable levels.
As anticipated from the particle size distribution, the Pedricktown soils generated a
small amount of fines (less than 1 percent of the original soil). This material was not
separately analyzed.
30
-------
These results indicate that AETS has the potential for remediating the contaminated
soil found on the Pedricktown site.
3.6 PALMERTON, PA SOIL
Site History
This soil is from the Superfund site in Palmerton, PA. The site is a mountain-side
adjacent to the Zinc Company of America, a zinc smelting company. Because of
deposition of zinc, cadmium, copper and lead, the mountain-side has become
completely defoliated. The soil was chosen for testing because of high levels of zinc,
and also because it contained some lead and cadmium. This soil is RCRA hazardous
because of the cadmium level.
Soil Description
The particle size distribution is given in Figure 10. The Figure represents 15 percent
of the soil, and so approximately 6 percent is under 20 microns. The soil was loam
and sand, with some clays and large particles (>i/s") present. The soil was also
blackish-brown in color with some visible yellow clays, tree bark, and vegetation. The
soil is well-weathered, and most of the fines and organic content have been washed
out over the years.
Figure 10 Palmerton Soil Particle Size Distribution
31
-------
Discussion of Results
Only one extraction was performed using this soil because the project possessed only
limited quantities. TCLP and total metals concentrations before and after the
extraction are summarized in Table 23.
The soil was hazardous because of the cadmium TCLP levels. AETS reduced the
cadmium TCLP value well below the limit. Zinc TCLP values were also reduced.
Copper values remained about the same, and lead actually increased. This indicates
that the extraction partially solubilized the lead, but did not efficiently remove it. Lead
in this soil is expected to be in the form of sulfides, which are not easily solubilized.
The data in Table 23 shows that the only metal in excess of the federal limits was
lead (over 500 mg/kg). Two metals exceeded the California total metals limits: zinc
and cadmium. Extraction using AETS reduced the total zinc concentration by over 90
percent, and the residual was well within the prescribed limits. The extraction
produced a similar 90 percent drop in cadmium levels, which were well within the
prescribed limits. The total lead concentrations were reduced to very near, but still
above, the federal limits of 500 mg/kg for surface soils, but below the concentrations
for subsurface soils and below the California total lead requirement. Total copper
concentrations were also reduced.
Table 23 AE-108: Palmerton Soil Experiment
Metal Removal Efficiencies from the Soil
Metal
Cd
Pb
Cu
Zn
TCLP (mg/L)
Untreated
2.60
0.66
0.16
70.5
Screen
0.17
2.03
0.15
4.26
Tank
0.25
3.68
0.23
3.76
% Removal
Screen
93.46%
N/A
6.25%
93.96%
Tank
90.38%
N/A
N/A
94.67%
Metal
Cd
Pb
Cu
Zn
Total Metals (mg/kg)
Untreated Screen Tank
137.0
898.0
166.0
9,150.0
11.6
844.0
93.7
707.0
9.0
588.0
54.9
352.0
% Removal
Screen
91.53%
6.01%
43.55%
92.27%
Tank
93.46%
34.52%
66.93%
96.15%
During the extraction, some of the soil was rinsed in the tank and some was rinsed
on the screen. When rinsing on the screen, the soil was moved across a
shaker/screen while being sprayed with the rinsate. The experiment demonstrated
that tank rinsing is the more thorough of the two methods for removing all of the
extractant.
32
-------
3.7 ACID REGENERATION
Regeneration of the acid figured prominently in the success of the extraction system.
In general, the reductions obtained from the pilot system exceeded those obtained
during the first year, laboratory-scale experiments. One explanation for this is during
the first year experiments, the extractant would tend to become saturated in metal
chlorides, and thereafter no metals would be dissolved. By attaching a regeneration
system which removes the metals and reforms the acid directly to the extraction tank,
the metals concentration in the extractant solution never builds up sufficiently to
approach equilibrium.
Specifics of the acid regeneration system are considered proprietary. Analyses of the
inlet and outlet extraction concentrations were conducted for nearly every extraction.
Results from two of the analyses are given in Table 24 and 25.
Table 24 Regeneration System Metal Removals for AE-105
First Extraction
Second Extraction
Metal
Cd
Pb
Cu
Zn
As
Cr
Ni
| in (mg/L) 1 out (mg/L)
2.561 0.50
50.2|| 2.1 2i
533" 4.77\
2780JI 11.2
58.7! 29.4J
59.2|| 0.47(
56' 0.55
% removed
80.5%
95.8%
99.1%
99.6%
49.9%
99.2%'
99.0%,
in (mg/L)
65.3
338
589
1580
N/A
N/A
N/A
lout (mg/L)
f 59
107
457
. 1490
N/A
N/A
N/A
I% removed!
9.6%,
68.3%!
22.4%
! 5.7%
. N/A
! N/A
N/A]
*N/A means the metal was not analyzed for.
NOTE: 'in' and 'out' refer to the concentration of total metals sampled as they were
going into and coming out of the acid regeneration system.
Table 25 Regeneration System Metals Removal for AE-107
Metal
Pb
cu
First Extraction
i n (mg/L)
1200
299
out (mg/L)
78.2
5.4
% removed
93.5%
8 98.2%
Second Extraction
in (mg/L)
975
325
out (mg/L)
88.3
404
% removed
90.9%
- 24.3%
NOTE: 'in' and 'out' refer to the concentration of total metals sampled as they were
going into and coming out of the acid regeneration system.
The results show that the acid regeneration system was capable of removing the
majority of metals from the acid stream. Results from two-stage extractions,
however, do show that the removal efficiency tended to decrease during the second
extraction. From the efficiency information, the required size of the regeneration
-------
system may be calculated based on the size of the regenerated stream and the metals
concentration within.
The removal efficiencies for arsenic in the acid regeneration system were lower than
that typical for other metals. This may be attributable to the form of arsenic ion -
typically AsO4~3 - which differs from the typically cationic form of other metals.
It is anticipated that for soils with high concentrations of metals, such as the
Pedricktown or SSM soils, a larger extractant regeneration system would be required
than for soils with relatively low concentrations of contaminants.
3.8 SOIL POST-TREATMENT
It is anticipated that the soils treated by AETS will eventually be returned to the
ground from which they were taken. The acid used during AETS will undoubtedly
remove some of the natural soil alkalinity, and produce a soil with depleted calcium,
and other nutrients. Therefore, it is anticipated that lime and fertilizer will be added
to the soil before placing it into the ground.
The addition of lime is anticipated to decrease the TCLP values for the soil. Therefore,
in order to conduct conservative tests which would tend to overestimate rather than
underestimate the TCLP in the resultant soils, CHMR did not add any lime or nutrients
to the soil after the extractions.
To determine the overall condition of the soils, CHMR had several samples analyzed
at an agricultural laboratory. The results from these analyses show that the soil pH's
tended to increase during the rinse step and after the extraction. Measured pH's of
treated soils varied between 4.9 and 6.6. Soil calcium levels also varied, from above
the optimum range for Pedricktown soils after treatment, to very low for the Butte
soil. Soluble salts levels tended to be within the normal range for soils.
Concentrations of phosphates, potash (potassium oxide), and magnesium tended to
be low after treatment in AETS. The results suggest that the soils will support
growth, after addition of small quantities of lime and nutrients.
3.9 FINES TREATMENT
CHMR performed additional laboratory-scale experiments to determine the treatability of
the fines fractions of the soil. At the time of this report, the results from only two
experiments were available. The first experiment, conducted with the fines from the
Pedricktown soil, showed a reduction of lead concentrations in a fines:water slurry from
20,300 mg/kg to 2,300 mg/kg. A second experiment conducted with a soil from a battery
breaker site in North Carolina showed reductions from 1,750 mg/kg to 550 mg/kg in the
fines: water slurry.
34
-------
Based on this positive result, CHMR is continuing investigations to develop a simple
treatment system for the fines fraction. It is anticipated that such a treatment system
will be applicable to both the fines fractions in soils and possibly sediment treatment.
35
-------
4.0 CONCLUSIONS
The results indicate that the Acid Extraction Treatment System is capable of treating
a wide variety of metals present in many different types of soil. Additional
conclusions may be drawn regarding the treatment system design, and specific
aspects of the technology. These are discussed in subsequent sections. In addition,
a final treatment system design is presented, and an economic analysis of it is given.
4.1 SOIL AND METAL TREATABILITY
Table 26 summarizes the soil treatability across the soils and metals tested. Where
individual soil fractions were separated during the extraction, and analyzed separately,
the table shows the composite results if the entire soil had been remixed. The results
show that AETS treated virtually all the soils tested to both reduce the total metals
concentrations to below currently regulated concentrations and reduce the TCLP to
below the currently regulated concentrations. Major exceptions included cadmium,
which consistently failed the TCLP for SSM soil, and lead, which failed both the TCLP
and total metals requirements for SSM soils. Lead was also not reduced below the
EPA surface soils concentration (500 mg/kg) for the Pedricktown soil, although it was
reduced below the EPA subsurface and California total metals concentrations.
Table 26 Qualitative Results of Extractions
Metal
As
Cd
Cr
Cu
Ni
Pb
Zn
Soil
SSM
*, T,L
*,T
*,T, L
*,T, L
*,T,L
*
*.T. L
Butts
*,T,L
*,T, L
*,T, L
*.T. L
King of Prussia
*,T,L
*,T,L
*,T, L
*,T, L
Pedricktcwn
*,T,L
*, T, L
*, T, L
Palmenon
*,T,L
*, T, L
*,T,L
*,T,L
Key: * ~ Metal is present in that soil
T- - Successful treatment for total metals
L- - Reduction in leachability to below standards.
Boldface and larger fonts indicate high initial
concentration (at least double the regulatory standards)
The total lead result for the Pedricktown soil is not surprising: the soil started with
nearly 3 percent lead, which was reduced to approximately 0.1 percent during a
single-step extraction. A second extraction is probably necessary to reduce the total
concentration further.
36
-------
Neither the total lead concentrations nor the TCLP lead concentrations were reduced
to below the regulatory limits during the extractions using the entire SSM soil. They
were reduced, however, during earlier extractions using only the coarse fraction of the
soil. Based on the results of the extractions, it appears that if the finest 20 percent
of the soil had been removed, the remaining soil would have been treatable using
AETS. CHMR is currently performing some experiments using the fines fractions to
determine if alternative extraction procedures could be used to remove lead from the
fine fractions of the soil.
The cadmium result for SSM is extraordinary in that the portion of the soil which
failed the TCLP for cadmium was the coarse fraction (+100 mesh). The fines passed
the TCLP. Most probably, the TCLP level in the coarse fraction could be reduced if
a longer extraction time were used. The Palmerton soil demonstrates that cadmium
is treatable using AETS - the soil is hazardous only because of the TCLP value for
cadmium, and that was reduced by over 90 percent using AETS.
4.2 AETS PROCESS DESIGN
Based on the results of the experiments, the basic design of the AETS process is
unchanged from that shown in Figure 1. The results have provided further information
and clarification concerning the following aspects of the required design.
Soil Pretreatment System
Extraction System
Dewatering and Rinse Systems
Acid Regeneration System
Post-treatment System
These components are discussed below.
Soil Pretreatment System
The soil pretreatment system is shown in Figure 11. The soil is first passed through
a grizzly, designed to remove particles larger than about 1 by 2 inches in size. The
underflow from the grizzly passes directly into an attrition scrubber, which is operated
at relatively high solids to liquid ratios. If makeup liquid is required in the scrubber,
it may be'supplied as regenerated acid from the extraction system. In the final section
of the scrubber, more liquid may be added if necessary to further slurry the soil and
make it easier to sieve.
Once through the scrubber, the soil passes directly onto a 6 mm wet screen. The wet
screen is sprayed with acid from the regeneration or extraction systems. The
overflow from the screen includes coarse gravel and bits of trees and other material,
which will be allowed to drain on a pad, and may possibly be rinsed to remove excess
small particles clinging to larger ones. The drainage from the coarse particles will
either be passed directly back onto the screen, or (if it is rinsewater), may be clarified,
treated to remove metals, and reused.
37
-------
SPHATW 10 WNSC ANO WE uovc tines
ATTRITION SCRU88IS
70S SOUQ5
Figure 11 AETS Pre-treatment
The underflow from the 6 mm screen will be placed directly into the extraction tank
for further processing.
The very coarse materials originally removed in the grizzly may or may not be
contaminated. If they are, they may be rinsed with the coarse particles (if necessary
to remove dirt and other clinging debris), or washed using debris washing techniques.
If they are uncontaminated, they can be returned directly to the site.
Extraction System
The extraction system consists primarily of a tank, or series of tanks, which provide
the soil with the appropriate contact time with acid. The tanks are well-mixed to
prevent solids from settling during the extraction. The soil is extracted at
approximately 20 percent by weight solids.
Figure 12 shows one possible system. The soils are fed to two extraction tank
system in series, which are intended to overcome problems associated with the
residence time distribution in continuously stirred tank reactors. Additional residence
time will be provided by the scrubber and sieving systems.
The soil slurry passes from the first tank to the second, then to a hydrocyclone. The
pilot plant used 2-inch hydrocyclones. Subsequent discussions with vendors and
experts on hydrocycloning suggest that a 4-inch cyclone may provide a better split.
38
-------
5C*«S 9OU0S
OEWATEMNO SCREEN
AND mac.
,f mes TO IAUK a
wv TO stew
f O I A*m 2
TANK 2
3 210 MINUTES
Figure 12 Extraction System
Therefore, the system is drawn using 4-inch cyclones. These will be manifolded
together as required for the system flow. Both the overhead and underflow from the
cyclone may be split, depending on the requirements of the individual soil. A portion
of the overflow may pass, directly into Tank 1, depending on the capacity of the clay
dewatering and acid regeneration systems. It is anticipated that sometimes,
particularly during start up, a portion or all of the underflow may also be cycled back
to Tank 1 and/or Tank 2. Otherwise, the underflow will pass to the coarse solids
dewatering and rinse system.
The overflow, meanwhile, will be clarified to remove clays and other solids. Then the
acid extractant will be regenerated. It is anticipated that approximately one-half of the
regenerated extractant will be passed directly back into Tank 1. The remaining
portion will be split among the scrubber (~ 10 percent), the sieve system (25 percent),
and Tank 2 (15 percent).
Rinse and Dewatering Systems
AETS is anticipated to require two rinse and four dewatering systems. Rinses will be
required for both the fine and coarse solids. Dewatering will be required of both the
fines and coarse solids both before and after they are rinsed.
39
-------
A dewatering and rinse system for the coarse solids is shown in Figure 13. The
system consists of a 200 mesh dewatering sieve, followed by a rinse tank, and a
second sieve system, The underflow from the dewatering sieve, which will contain
some fines, is passed back to the extraction tank. The fines are anticipated to build
up in the extraction tank and to be removed with the sieved solids.
CVCIONE UNDERFLOW
Y
seve BOO MSSHI |
V
V V
w
T
SACK TGEA"KiCTtCN TANK
CLEAN SOU
10»O8T-TREATMENT
(A)
REMOVAS,
Figure 13 Coarse Solids Dewatering and Rinsing System
The overflow from the dewatering sieve is passed into a rinse tank, where it is rinsed
and well-mixed for 10 minutes or more. From there it passes to a second dewatering
sieve, also 200 mesh. The underflow from this sieve is flocculated and clarified. The
solids from the floe tank are carefully placed atop the dewatering sieve for removal
with the clean soil. The clarified rinsate is further processed to remove the metals and
then recycled to the rinse tank.
The fines dewatering and rinse system is shown in Figure 14. The fines are first
sieved to remove any particles larger than 100 mesh, which are returned to the
extraction tank. They are then sent to a clarification unit, and allowed to flocculate.
The clarified extractant is regenerated and returned to the extraction system. The
thickened clays are further treated (if necessary), then rinsed, and thickened again.
The clarified rinsate is treated to remove metals, then returned to the rinse system.
40
-------
The thickened, rinsed clays are further thickened in a filter press and sent to soil post-
treatment.
CTCIO« OVERFLOW
y
, 100MfSMS*VE
THICKENER
THICKENED curs '
10TOlS%SOUOS
»IOC MESH TO EXTRACTION TANKS
CLAftFCOBfTHACTANTTO
N£«A,TlON SYSTEM
FURTHER TfCATMENT
A
i\ y
\ -^
\ _
BlNSg TANK
FINES TO PQST-T
V
y
Figure 14 Fines Dewatering and Rinsing
Acid Regeneration System
AETS employs a proprietary acid regeneration system.
Soil Post-Treatment
The purpose of soil post-treatment is two-fold: to remix soil fractions which have
become separated during processing and to return the soil to its native condition.
The soils are anticipated to be remixed using front-end loaders or other earth-moving
equipment. The soils will be mixed with a small amount of lime to return the soil
buffering capacity. In addition, fertilizers and topsoil may be added. Experiments to
determine the exact post-treatment requirements are on-going.
It should be emphasized that the TCLP results to date have not included any post-
treatment. It is anticipated that the addition of lime, and other neutralizing agents will
41
-------
help stabilize the metals in the soils. In addition, any significant addition of topsoil or
fill will dilute both the TCLP and total metals results. Neither lime nor topsoil were
added during the experimental program, and their addition is not included in a
determination of whether or not the AETS treated soil is hazardous.
42
-------
5.0 AETS ECONOMICS
To estimate the economics of remediating a site using AETS, capital and operating
costs for various sized and configured systems were determined. These are
summarized in Table 27.
Table 27 AETS Cost Summaries Under Various Conditions
Process and Site Parameters
reed
Rate
yd'/hr
30'
20
20
20
15
15
15
10
Extraction
Res. Tim«
(min)
24
24
36
24
24
36
36
36
% Fines
(<50pm)
15
15
30
15
15
30
15
36
% of
Fines
Disposed
2
2
25
25
2
25
2
25
Metals
cone.
(mg/kg)
5,000
5,000
15.000
15,000
5.000
15.000
5.000
15.000
Site size
(1 000 yd3)
150
100
60
80
60
30
30
20
Costs
Capital
Costs
(million $)
4.5
3.6
4.5
4.1
3.2
3.3
3.3
3.2
Operating
Costs
($/yd3)
41
51
82
71
61
92
61
112
Total Cost perydj
Capital paid
off over one
site (5/yd3)
83
104
178
141
133
243
189
301
Capital paia
off over two
sites ($/yd3)
71
88
147
121
111
191
146
237
Notes: 1. Plant is anticipated to operate only 1 shift per day.
2 No metal recovery value is assumed; all metal sludges are disposed.
5.1 COST CALCULATIONS
Capital costs were calculated by summing the following:
General Costs
Pretreatment
Extraction Cost
Acid Regeneration
Dewater/Rinsing
Including site preparation, pilot work, trailers, and
permitting. These represent approximately 8 to 11
percent of total Capital Costs.
Costs associated with coarse and very coarse
removal, scrubbing, and coarse rinsing/processing.
(11 to 15 percent of total capital).
Costs associated with contacting the soils with acid,
including hydrocycloning the soil (7 to 9 percent)
Costs associated with metals removal and acid
reformation. (31 to 41 percent)
Costs associated with dewatering and rinsing coarse
solids and thickening and processing fines. (21 to 25
percent)
43
-------
Soil Post-treat Costs associated with mixing; post-treatment and
fertilization. (3 to 4 percent)
Miscellaneous Costs associated with other required piping
instruments, etc. (7 to 9 percent)
The costs for each element were increased by 10 percent for engineering, 23 percent
for transportation and final installation costs, 5 percent for start up and shakedown
costs, 2 percent for spare parts, and 10 percent for contingency. The most uncertain
costs were those determined for the regeneration system. To the extent possible,
these costs were overestimated, so that the final unit costs were conservatively
determined.
Operating costs included labor (a total of between 3 and 4 operators, plus a
supervisor, 2 to 4 excavators (with excavation equipment), a health and safety officer,
maintenance and engineering), maintenance equipment costs, utilities, chemicals,
disposal, and reseeding.
The capital and labor costs were combined by assuming a 10 percent cost of capital;
depreciation over either 1 or 2 years; operations 1 shifts per day (2000 hours/year)
for one year per plant site; moving and reassembly costs of $150,000 plus 15 percent
original capital (if the plant is depreciated over two years); and plant downtime of 10
percent of operating hours for unanticipated shutdowns (i.e., equipment failure).
5.2 COST SUMMARY
Table 27 gives a cost summary for AETS at several different process configurations.
The table shows the effects of varying six critical parameters (feed rate, extraction
time, percent fines, percent fines disposed (rather than treated), metals
concentrations, site size and the number of sites treated with each set of equipment).
Note that the table includes costs for mobilization, pilot plants, excavation, replacing
soil, and reseeding the ground as well as soil treatment. Thus, the costs represent the
total costs of treatment using the Acid Extraction Treatment System.
For 15 and 20 cubic yard per hour plants, the table gives the cost under best
conditions (first row), cost under worst conditions (second row), and cost under
intermediate conditions (third row). The table also gives the cost under best
conditions for the largest plant anticipated (30 yd3/hr), as well as the cost under worst
conditions for the smallest plant anticipated (10 yd3/hr). In this way, the table should
bracket the costs. For reasonably sized plants, the anticipated treatment costs range
between $100 and 180 per cubic yard.
44
-------
The costs for the most commonly employed alternative (stabilization and disposal),
range between $180 or $450 per cubic yard, depending on the size and
circumstances of the site, with typical costs about $250 per cubic yard. Thus, AETS
is generally competitive with stabilization and disposal. It is also a more
environmentally sound alternative because of the potential for reclaiming the metals
found at the site.
45
-------
APPENDIX A
EQUIPMENT LISTS
This appendix gives a detailed list of the equipment used in the Acid Extraction
Treatment System pilot-scale plant operation.
B. 1 MAJOR EQUIPMENT
This section describes all of the larger pieces of equipment that were used in the
actual extraction process.
Attrition Scrubber
Manufactured by Denver Corporation
Serial number 195-264-001
The scrubber was made custom for CHMR with dual 230 V, three-phase motors and
a special rubber lining to prevent corrosion from the acidic solution. The scrubber was
modified to be drained out of the cleaning port instead of the 6-inch flanged outlet due
to the batch application. Six casters were attached to the base of the scrubber for
safety purposes when moving the scrubber for cleaning, etc.
Slurry Pump
Manufactured by Galigher Corporation, A Division of BGA International
Model number 1.5 VRA 1000
Serial number 0248592
This pump was custom modified for the acid environment with which it was to be
used. The pump head was coated with a gum rubber lining with a Buna-N impeller.
The pump was driven by a 2 horsepower, 1800 RPM, 230 volt, three-phase motor,
The motor was connected to the pump by an adjustable V-belt drive. The pump
features a 21/2-inch inlet and a 2-inch outlet, both flanged. The sheaves, belts, and
motor for the pump were supplied by Allegheny Process Equipment.
Rubberized Pump
Manufactured by GRI
Model number 08107-002
46
-------
This pump was centrifugal with a rubber head to handle the acid. The pump
malfunctioned due to the abrasion, and was later repaired and used to pump to the
1-inch cyclone.
Two-Inch Cyclone
Manufactured by Krebs Engineering
Model number U2-1436
Serial number 48034
This was the primary cyclone for the AETS. This cyclone separated the coarse sand
from the clay fraction. The 2-inch cyclone featured an adjustable apex that was later
replaced by a fixed plastic apex. The cyclone had 11/4-inch threaded inlets and
outlets.
One-Inch Cyclone
Manufactured by Krebs Engineering
Model number PC1 -S312
Serial number 44869
This cyclone was used to dewater the clay from the 2-inch cyclone. It was also used
to separate the clay from the extractant or rinsate. The inlet and outlet to the cyclone
were 3/8-inch and 1-inch, respectively.
pH Controller
Cole-Parmer pH/ORP/CD Pump System Series 7142
Model number 7142-55
Serial number 22941
This controller featured a diaphragm pump and two-way acid/base operation. Once
calibrated, the controller needed no supervision or adjustments.
Extract Filter
Manufactured by Harmsco Industrial Filter
Model number H1F7
Serial number 6390
This was an up-flow type filter with the inlet coming from the extract pump, and the
outlet going to the acid regeneration system. The filter contained seven filter
cartridges, either 5 or 10 ^m in size.
47
-------
Extract Pump and Controller
Pump was manufactured by Sherwood
Model number CMNP-1231-W
E-TRAC "S" Type AC inverter/controller was manufactured by
T. B. Wood's Sons Company
Model number AFC-2000-7B2S
Serial number 05200-R-090
The speed of the centrifugal pump was varied by the digital controller, allowing the
pressure to the filter to be controlled.
Shaker Sieve and Screens
Manufactured by Liquatex Separators, Inc.
» Model number L22-1-0
Serial number LI80883
The 2 by 2 foot shaker was the most diverse piece of equipment used in the acid
extraction system, serving three purposes. First, it was used to separate the +8
mesh (>1/s") particles from the raw soil. Next, it was used to dewater the soil after
each extraction. Finally, the shaker was used to dewater the soil after the rinsing
stage.
8.2 MINOR EQUIPMENT AND PIPING
This sections describes the minor equipment and plumbing used during the course of
the research.
Transfer Pumps
Manufactured by March Inc.
Model number 1A-MD-1
Manufactured by Little Giant Pump Company
Model number 4E-34NR
Serial number YY-352-3272
These were submersible pumps used to transfer various liquids, such as extractant,
rinsates, etc., while preparing for or cleaning up after extractions.
48
-------
Rinse Regeneration Pump
Manufactured by Little Giant Pump Company
Model number LG-100
Serial number WW-2029603
This peristaltic pump was used to pump the rinsate to the rinsate regeneration
system.
Plumbing
Various piping was used to plumb the system together, including:
1 1/2" ID PVC pipe from the extraction tank to the slurry pump
1" ID Tygon tubing for the cyclone outlets
3/4M ID PVC pipe everywhere else
PVC fittings including:
- unions
tees
- 45° elbows
90° elbows
caps
- flanges
- all thread nipples
reducer bushings
bulkheads
ball valves
gate valves
gauge guards (with pressure gauges)
49
-------
APPENDIX B
QUALITY ASSURANCE/QUALITY CONTROL
As specified in the project Quality Assurance Project Plan (QAPP), the Quality
Assurance/Quality Control (QA/QC) procedures during the study incorporated two
levels:
» Analytical laboratory QA/QC procedures and checks; and,
» Process QA/QC procedures and checks.
The analytical QA/QC program was mainly concerned with the quality of data received
from the analytical laboratory. To this end, duplicate samples were used, split
samples were sent to two different laboratories, and blank samples were submitted
to the laboratory for analysis. In addition, the laboratory periodically performed
internal duplicate and spike analyses, as required in the analytical methods and in their
QA/QC programs.
Process QA/QC procedures and checks involved methods to check the data received
from the process experiments. To this end, duplicate experiments were performed
under identical conditions to determine if similar results were obtained.
The results from these QA/QC procedures and checks are summarized in the tables
below.
Analytical Laboratory QA/QC
The results from select duplicate and split samples from several analyses are
presented below. These results indicate that, in general, the relative percent
differences were within the required 25% for duplicate samples. However, some
discrepancies did occur. These were resolved typically by rechecking the data or
repeating the analyses. TCLP values in the table are in units of mg/l and total values
are given in units of mg/kg.
EXPERIMENT
AE-102
AE-102
AE-1 20
AE-120
METAL
Pb
Zn
As
Cd
ANALYSIS
(TOTAL/TCLP)
Total
Total
Total
Total
RESULT
#1
77
770
620
970
RESULT
#2
98
582
730
1,300
RELATIVE
PERCENT
DIFFERENCE.
21%
24%
15%
25%
50
-------
AE-120
AE- 120
AE-120
AE-120
AE-107
AE-105
AE-105
Cr
Pb
Ni
Cu
Pb
Cd
Ni
Total
Total
Total
Total
TCLP
TCLP
TCLP
1,300
10,040
980
10,900
520
8.29
11.6
1,320
12,300
1,410
10,600
503
8.36
11.8
1 %
18%
30%
3%
3%
1 %
2%
The analytical laboratory use of spikes and duplicates was reviewed, and found to
consistent with the prescribed QC checks.
Process QA/QC
In order to determine how consistently the overall process was performing, and also
to serve as a QC check on the overall combined process, laboratory analytical, and
sampling procedures, CHMR duplicated several entire experiments. These
experiments included runs AE-102 and 104; AE-105 and 106; and AE-115 and 116.
Typical results from these duplications are given in the table below, which shows the
results from the AE-115 and 116 experiments.
Metal
Cr
Cu
Ni
INITIAL SOIL
TOTAL METALS
(mg/kg)
AE-115
1020
1240
335
AE-116
1240
1660
518
RPD
22%
25%
35%
FINAL SOILS
TOTAL METALS
(mg/kg)
AE-115
37
17
4
AE-116
89
18
6
RPD
58%
6%
33%
The results from these analyses indicate a reasonable degree of consistency between
runs, with average relative percent differences of less than 25 percent.
51
------- |